English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Accurate prediction of aerodynamic forces in real-time is crucial forautonomous navigation of unmanned aerial vehicles (UAVs). This paper presents adata-driven aerodynamic force prediction model based on a small number ofpressure sensors located on the surface of UAV. The model is built on a linearterm that can make a reasonably accurate prediction and a nonlinear correctionfor accuracy improvement. The linear term is based on a reduced basisreconstruction of the surface pressure distribution, where the basis isextracted from numerical simulation data and the basis coefficients aredetermined by solving linear pressure reconstruction equations at a set ofsensor locations. Sensor placement is optimized using the discrete empiricalinterpolation method (DEIM). Aerodynamic forces are computed by integrating thereconstructed surface pressure distribution. The nonlinear term is anartificial neural network (NN) that is trained to bridge the gap between theground truth and the DEIM prediction, especially in the scenario where the DEIMmodel is constructed from simulation data with limited fidelity. A largenetwork is not necessary for accurate correction as the linear model alreadycaptures the main dynamics of the surface pressure field, thus yielding anefficient DEIM+NN aerodynamic force prediction model. The model is tested onnumerical and experimental dynamic stall data of a 2D NACA0015 airfoil, andnumerical simulation data of dynamic stall of a 3D drone. Numerical resultsdemonstrate that the machine learning enhanced model can make fast and accuratepredictions of aerodynamic forces using only a few pressure sensors, even forthe NACA0015 case in which the simulations do not agree well with the windtunnel experiments. Furthermore, the model is robust to noise.

Machine learning enhanced real-time aerodynamic forces prediction based on sparse pressure sensor inputs